Colour distributions in E-S0 galaxies · 246 R. Richard: Colour distributions in E-S0 galaxies. IV. deemed useful to analyse the best B and R frames in Nieto’s CFHT collection,

ASTRONOMY & ASTROPHYSICS JUNE I 1999, PAGE 245

SUPPLEMENT SERIES

Astron. Astrophys. Suppl. Ser. 137, 245–268 (1999)

Colour distributions in E-S0 galaxies

IV. Colour data and dust in E’s from Nieto’s B, R frames?

R. Michard

1 Observatoire de la Cote d’Azur, Dept. Augustin Fresnel, BP. 229, F-06304 Nice Cedex 4, France2 Observatoire de Paris, DEMIRM, 77 Av. Denfert-Rochereau, F-75015 Paris, France

Received October 23, 1998; accepted March 9, 1999

Abstract. The B − R colours distributions (with R inCousins’s system) have been measured in 44 E classifiedgalaxies in the Local Supercluster, from pairs of framescollected by Nieto and co-workers in 1989-91. These arenearly all from the CFHT, and of sub-arsec resolution.

Great attention has been given to the effects of un-equal PSF’s in the B and R frames upon colour distribu-tions near centre; such effects are illustrated from modelcalculations and from pseudo-colours obtained from pairsof frames taken in the same band but with different see-ing conditions. Appropriate corrections were systemati-cally applied in order to derive central colours and innergradients, although still affected by the limited resolutionof the frames.

The radial colour distributions have been measured inmore detail than usual, considering separately the nearmajor axis and near minor axis regions of the isophotalcontours. Azimuthal colour distributions, in rings limitedby selected isophotes, were also obtained.

Dust “patterns”, i.e. patches, lanes, arcs, ..., have beendetected and mapped from the colour distributions. An adhoc dust pattern importance index (or DPII) in a scaleof 0 to 3, has been introduced to qualify their size andcontrast.

We have tried to find evidence of a diffuse dust con-centration towards the disk, if one is apparent. Positiveresults (noted by the dd symbol) have been obtained fordisky E’s, whenever the inclination of their disk to the lineof sight is large enough, and eventually also in the smallisolated disks sometimes present in both boxy and diskygalaxies.

The red central peak occurring in many E-galaxiesmight be the signature of a central concentration of dust,also in cases where this peak is isolated rather than em-bedded in some extensive colour pattern. The properties

Send offprint requests to: R. Michard? Based on observations collected at the Canada-France-

Hawaii Telescope and at the Observatoire du Pic du Midi.

of the near centre colour profiles have been related toa classification of nuclear photometric profiles into “flattopped” and “sharply peaked” (equivalent to “core-like”and “power-law” in the terminology of Faber et al. 1997).

The published here data include the following:. Short descriptions and codes for the characters of the

B −R distribution of each object, and comparison to theresults of recent surveys.

. A table of the mean B −R at the centre and at twoselected isophotes, a “core colour gradient” and the usuallogarithmic gradient.

. Maps of near core B−R isochromes and B isophotesfor comparison.Images of the B−R colour distribution are made availablein electronic form.

Key words: galaxies: elliptical and lenticulars, CD —galaxies: ISM — galaxies: fundamental parameters

1. Introduction

In 1989-91, the late J.-L. Nieto and co-workers obtainedat the CFHT frames of many elliptical galaxies, plus a fewS0 objects, in order to investigate the core profiles of theseobjects. Preliminary results were published by Nieto et al.(1991a, 1991b). They found that the cores of E-galaxiescould be empirically sorted out into two classes, one wellor nearly resolved at the CFHT resolution, the otherremaining quite unresolved: the disky ellipticals (or diE),like the S0’s, were in the sharp nucleus class. The problemof galaxian central profiles has now been tackled success-fully by several groups using HST frames, with specialemphasis upon E-S0 galaxies, providing more quantitativeconfirmations of the conjectures of Nieto’s et al. Howevercore colour profiles and patterns of such objects have notyet been much studied at high resolution. It was therefore

246 R. Richard: Colour distributions in E-S0 galaxies. IV.

deemed useful to analyse the best B and R frames inNieto’s CFHT collection, restricting the sample torelatively nearby galaxies, i.e. objects in the LocalSupercluster with V0 < 3000 km s−1. Suitable pairs forour goal were available for 38 E-type galaxies, that isobjects so classified in one of the usual catalogues. Thedata was supplemented by Pic du Midi B, R frames for 6E objects, obtained by E. Davoust and kindly put at ourdisposal.

In previous papers of this series some results werepresented, based largely or exclusively upon the present“Nieto’s sample”. Such results are shortly summarized be-low. These papers also described the background and pur-poses of our work. The corresponding comments and refer-ences will not be repeated here. Review papers of interesthave been published by Goudfrooij (1995) and (1996).

Our previous results are here recalled:

1. In Paper I of this series (Michard 1998a) a dust patternimportance index (or DPII) was obtained for a nearlycomplete sample of 67 ellipticals, using the presentNieto’s sample supplemented by literature data. Thefrequency and importance of dust patterns for vari-ous brands of E-classified galaxies were studied fromstatistics of this index. It was found that dust is muchmore frequently detected, and with more importantpatterns, among diE’s than boE’s, the unE’s beeingintermediate.

2. In Paper II (Michard 1998b) evidence for dustconcentration in the disk was sought, for galaxiescontaining such a component: the criteria were anasymmetry in light and colour along the minor axis,as described for S0’s by Michard & Simien (1993)(MS93). and/or a flattening of the isochromes as com-pared to the isophotes. In case of positive evidence,the symbol dd for “dust in disk” was introduced. Thisoccurs in diE’s, whenever the inclination of their diskto the line of sight is large enough, and eventally alsoin the small isolated disks sometimes present in bothboE’s and diE’s galaxies.

3. In Paper III (Michard 1998c) the colour amplitude ofthe central red peak detected in most E-galaxies wasdescribed by the parameter ∆C0,3, i.e. the difference ofcolour between the centre and the isophote of mean ra-dius 3 arcsec. For a sample of 39 galaxies, the centralprofiles were classified into two classes, shown to beequivalent to the “core-like” and “power-law” types ofFaber et al. (1997). It was found that the red peak am-plitude is very small for “core-like” objects, unless theyshow clear evidence of a dust pattern. This agrees withrecent HST results by Carollo et al. (1997), from a spe-cial subsample of objects with kinematically decoupledcores, and possibly rule out a model for the formationof flat cores proposed by Silva & Wise (1996).

This paper will present our data in more detail than in thepreviously published papers. Section 2 will describe the

observations. Section 3 will introduce the techniques to beused in order to derive reliable 1D and 2D colour distribu-tions. A special emphasis is given to errors induced by thedifferent PSF’s of the two frames involved in colour mea-surements, and to the techniques applied to reduce sucherrors. Section 4 describes various typical features of thecolour distributions in E-galaxies, and introduces usefulad hoc parameters. Section 5 compiles the data of interestfor statistical discussions and presents comparisons withother works.

Further papers of this series will deal with:

. An examination of colour distributions in stronglyinclined S0 galaxies, a project started in cooperation withDr. P. Poulain in Toulouse. The available material for thiskind of objects is rather scanty, except for the study ofNGC 3115 by Silva et al. (1989), and the observations ofbulges by Balcells & Peletier (1994).

. Statistics of the colour material gathered atObservatoire de Haute-Provence in collaboration withthe late J. Marchal.

1.1. Notations and abbreviations

It may be useful to collect here the notations and abbre-viations currently used below.

. SuBr surface brightness.

. majA, minA, major and minor axis respectively

. a, c major and minor axis of the Reference Ellipsefor the representation of an isophote according to Carter(1978).

. r = (ac)1/2 mean radius of the isophote.

. ei, fi coefficients of cosine and sine terms in Carter’sharmonic representation of deviations from the ReferenceEllipse.

. PA position angle of major axis.

. q axis ratio of the Reference Ellipse.

. ε = 1− q ellipticity.

. diE, boE, unE subclassification of ellipticals as disky,boxy or undeterminate; p added for peculiar envelopes.

2. Observations

Two set of observations are used in this survey. The firstis from 4 runs at the CFHT in 1989-91, the second fromone night at the TBL (Telescope Bernard Lyot) of the Picdu Midi. Tables 1 and 2 list the used frames with somerelevant informations. Since the frames were intended tomeasure central light profiles, the exposure times weregenerally short, and many frames are of rather poor S/Nratio. A classification of S/N in terms of G (good), F (fair)and P (poor) is given in the table.

R. Richard: Colour distributions in E-S0 galaxies. IV. 247

Table 1. List of Cassegrain CFHT frames used in this work.F = filter. Exposure times are in seconds. W = FWHM inarcsec. Code for S/N : G = good; F = fair P = poor. R = ap-proximate range of measured radius in arcsec. When the filecode contains an m, it was obtained by merging two frames.Individual notes: NGC 0821: nearby bright star. NGC 3115:Frames used here for technical tests

NGC F File Date Exp W S/N R

0584 B h051 05/12/89 180 1.15 G 30R h051 id 60 1.02 G 30

0596 B h051 05/12/89 210 1.10 G 30R h051 id 90 0.75 G 30

0636 B h051 05/12/89 300 1.13 G 30R h051 id 150 0.96 G 30

0720 B h051 05/12/89 180 - G 20R h051 id 60 - G 20

0821 B h051 05/12/89 300 0.90 G 20R h051 id 120 0.87 G 20

1052 B h051 05/12/89 180 0.96 G 25R h051 id 60 0.82 G 25

2768 B h062 06/12/89 300 1.00 G 20R h063 id 120 0.66 G 20

2974 B h062 06/12/89 120 0.68 F 20R h06m id 90 0.68 G 20

3115 R h051 05/12/89 30 0.91 F 40R h064 06/12/89 60 0.55 G 35

3377 B h052 05/12/89 150 - G 40R h052 id 150 - G 40B h061 06/12/89 45 0.76 F 20R h062 id 15 0.59 F 20R h063 id 15 0.66 F 20

3379 B h06m 06/12/89 60 0.81 F 30R h06m id 30 0.60 F 30

3585 B h052 05/12/89 180 1.16 G 30R h052 id 90 1.07 G 30

3610 B h061 06/12/89 90 0.72 F 20R h061 id 15 0.52 P 20R h062 id 20 0.59 P 20R h063 id 60 0.60 F 20

3613 B h061 06/12/89 30 0.80 P 10R h061 id 90 0.55 F 10

4125 B h042 04/04/90 120 0.91 F 20R h042 id 45 0.77 F 20

4365 B h042 04/04/90 120 0.82 F 25R h042 id 45 0.96 F 25

4374 B h041 04/04/90 120 - F 30R h042 id 15 - F 30

4387 B h042 04/04/90 120 1.03 F 20R h042 id 60 0.70 G 20R h043 id 120 0.69 G 20

4406 B h041 04/04/90 120 0.67 F 20R h041 id 45 0.74 F 20

4472 B h042 04/04/90 120 0.77 F 30R h042 id 45 0.72 F 30

4473 B h04m 04/04/90 135 - F 20R h042 id 45 - F 20

4478 B h043 04/04/90 300 0.87 F 20R h042 id 45 0.71 F 20

4494 B h04m 04/04/90 135 1.20 F 20R h041 id 45 0.75 F 20

Table 2. List of HRCam CFHT frames used in this work. Seeprevious table for explanations. Individual notes: NGC 4660frame h073: a defect near the galaxian core could be correctedusing another frame of lesser S/N

NGC F File Date Exp W S/N R4261 B h202 20/04/91 300 0.77 P 25

R h202 id 60 0.80 P 254406 B h202 20/04/91 300 0.85 F 35

R h202 id 120 0.72 F 35R h203 id 90 0.61 F 35

4472 B h202 20/04/91 300 0.96 F 25B h203 id 300 0.82 F 25R h202 id 120 0.86 F 25R h204 id 90 0.72 F 25

4494 B h202 20/04/91 300 0.89 F 30R h202 id 60 0.77 F 30

4551 B h212 21/04/91 240 - P 20R h212 id 90 - P 20

4564 B h212 21/04/91 180 - F 25R h212 id 90 - F 25

4621 B h203 21/04/91 210 0.81 F 15R h203 id 30 0.85 F 15

4742 B h192 19/04/91 240 0.77 F 15R h191 id 45 0.61 F 15

5322 B h202 20/04/91 480 0.89 F 15R h204 id 150 0.78 F 15

5813 B h20m 20/04/91 450 0.84 F 15R h203 id 150 0.75 F 15

5845 B h212 21/04/91 120 0.66 P 10R h213 id 60 0.48 P 10

5846 B h212 21/04/91 120 0.70 P 15R h213 id 15 0.61 P 15

5846A B h21m 21/04/91 210 0.70 P 8R h214 id 120 0.61 F 8R h216 id 45 0.64 P 8

4478 B h072 07/06/91 180 0.62 P 20R 072 id 60 0.50 F 20

4486B B h072 07/06/91 120 0.71 P 6R h072 id 30 0.57 P 6

4649 B h072 07/06/91 240 0.69 F 10R h072 id 150 0.65 F 10

4660 B h073 07/06/91 180 0.81 F 15B h07m id 180 0.81 F 15R h073 id 60 0.73 F 15

5576 B h072 07/06/91 180 0.99 F 15R h072 id 60 0.60 F 15

5638 B h072 07/06/91 180 0.90 P 15R h072 id 120 0.69 F 15

5831 B h07m 07/06/91 180 0.97 P 20R h062 id 90 0.60 F 20

2.1. The CFHT observations

The CCD camera at the Cassegrain focus was used in theruns of December 1989 and April 1990. It gives a scale of0.107 arcsec/pixel, with a good sampling even at the bestseeing. For the two runs of 1991, the so called HRCamwas used at the Prime focus: it compensates for telescopeguiding errors using an auxiliary star in the frame field,with some improvement of the seeing. The image scale ispractically the same as for the Cassegrain camera. Sincethe effective entrance aperture for HRCam is reduced to120 cm, the S/N is much less good, at equal exposuretime, than with the original Cassegrain camera.

The used frames are nearly all of sub-arcsec resolution.A small statistics of mesured stellar FWHM is given inTable 4. The overall means are 0.72 arcsec in R and 0.86 in


Table 3. List of Pic du Midi frames used in this work. See pre-vious table for explanations

NGC F File Date Exp W S/N R

3156 B d201 20/03/90 900 1.13 P 35R d201 id 300 0.98 P 35

3193 B d201 id 600 1.16 P 30R d201 id 180 0.82 P 30

3605 B d201 id 600 1.05 P 20R d201 id 180 1.08 P 20

3608 B d202 id 600 0.82 P 35R d201 id 180 0.83 P 35

3640 B d211 id 600 - P 35R d211 id 180 - P 35

3872 B d212 id 600 1.32 P 30R d211 id 180 0.92 P 30

Table 4. Statistics of measured stellar FWHM ’s at the CFHT.Night: date of night beginning in Hawai time. N : number ofused observations for each filter. WB: mean FWHM in B (arc-sec). σB : standard deviation. WR: mean FWHM in R. σR:standard deviation

Night N WB σB WR σR05/12/89 6 1.04 0.11 0.94 0.1506/12/89 8 0.79 0.10 0.60 0.0504/04/90 7 0.89 0.18 0.76 0.1019/04/91 1 0.77 - 0.61 -20/04/91 7 0.83 0.08 0.77 0.0921/04/91 3 0.69 0.03 0.57 0.0806/06/91 7 0.81 0.15 0.65 0.10

B. On the other hand, the usable field is 641×1011 pixels,or only 68× 108 arcsec, so that the determination of thesky level is rather crude. In order to get a star suitable forPSF measurement within the field, it was often necessaryto position the galaxy well away from the frame centre,with further reduction of the measurable range. Due tospecific technical constraints this difficulty is more acutewith the HRCam.

2.2. The TBL observations

The observations for 6 galaxies are from a single goodnight (see Table 3). The camera, at the then used focalreducer, gives a scale of 0.315 arcsec/pixel, with adequatesampling for the encountered seeing. For the frames un-der study, the average stellar FWHM are 0.92 arcsec inR with σ = 0.08 and 1.13 in B with σ = 0.20. Their fieldis 107× 174 arcsec.

3. Data analysis

3.1. Preliminaries

The data analysis involves the following preliminary steps:

1. The usual correction of the frames for instrumentaleffects were made by the observers, following the rou-tines in use at each observatory.

Fig. 1. Sample of a measured radial B−R distribution, i.e. forNGC 4125. Open circles: eastwards majA. Filled circles: west-wards majA. Open squares: southwards minA. Filled squares:northwards minA. The differences are due to an important dustpattern, the data being smoothed by averaging along arcs ofisophotes as explained in the text. The upper label gives thecode of the B-frame and the corresponding seeing FWHM

Fig. 2. Sample of a measured azimuthal B − R distribution,i.e. for NGC 4125. Four successive “rings” (as defined in3.2.1), limited with the isophotes of radii 2.0, 3.2, 5.0, 7.9 and12.6 arcsec are measured. Abscissae: PA along the isophotescounted counterclokwise from one of the majA, here the onepointing westwards. Ordinate: B − R, the scale for each ringbeing adjusted as needed. The large variations are due to thedust pattern. The upper label gives the code of the B frameand the corresponding seeing FWHM


Fig. 3. Experiment showing a pseudo-colour profile induced bydifferential seeing, and its approximate correction. Abscissae:log of isophotal mean radius r in arcsec. Ordinates: Colour inmagnitude, with circles for the majA and crosses for the minA.The model galaxy is a circular r1/4 bulge (slightly modified),of FWHM = 0.60 arcsec. The PSF’s are sum of Gaussians.Sharper one DW31: FWHM = 0.51 arcsec with faint wings.Broader one DW43: FWHM = 0.79 arcsec with strong wings.Upper curve: Colour profile for convolved frames of the model.Lower curve: Colour profile after PSF matching with a sin-gle circular gausian (case CG1R). The correction is quitesuccessful

2. The preparation of each frame for measurement wasmade in the ESO-MIDAS environment, using a pro-cedure summarized by Giudicelli & Michard (1993).It involves the elimination of significant parasitic ob-jects, the extraction of a suitable stellar image (if any)for the derivation of the PSF, the evaluation of thesky background, the calibration by comparison withthe results of aperture photometry (see Sect. 3.3 be-low); a cosmetic treatment against cosmic ray peaksand anomalous negative pixels, and reduction of theframe to the field deemed necessary. If the S/N ratiois adequate for the derivation of 2D colour maps, theabove treatment may be preceded or completed by themutual alignement of the frames against the one takenas geometrical reference, possibly using the images ofstars to improve the alignment. The output of thesepreparations are “clean” frames for the galaxy and thePSF, with a number of useful parameters available inthe frames “descriptors”.

3. Since the colours will be measured along arcs ofisophotes, it is a necessary step to obtain a set ofisophotes in tabular form. The contours are describedby the well known representation first proposed by

Fig. 4. Experiment showing pseudo-colour profiles induced bydifferential seeing, and their approximate correction. Abscissae:log of isophotal mean radius r in arcsec. Ordinates: Colourin magnitude, with circles for the majA and crosses for theminA. The model galaxy is akin to a flat disky E or S0, witha slightly modified r1/4 bulge of ellipticity ε = 0.5, plus a diskwith ε = 0.74. The PSF’s are sum of Gaussians, i.e. the sameas in Fig. 3. Upper curve: Colour profiles for convolved framesof the model. Lower curve: Colour profiles after PSF matchingwith a single circular gausian (case CG1R). The differentialseeing results in a blue light excess on the minA, extending upto 3 − 4 times the broad PSF FWHM . Again the correctionis successful

Carter (1978). For details about our implementationof Carter’s representation, see Michard & Marchal(1994). Only one set of isophotal contours is used tocompare two frames and get local colours, but thecomparison of Carter’s parameters for the two framesmay be revealing, as noted for instance by Goodfroijet al. (1994a).

3.2. Colour measurements

In previous survey of elliptical galaxies, one was often sat-isfied with measuring a local reference colour, plus a colourgradient, such as C1 − C2 and d(C1 − C2)/dlog r, wherer is the equivalent isophotal radius defined above. In gen-eral, however, two parameters are far from sufficient todescribe the colour distribution in an early-type galaxy.There is no a priori physical reason for the logarith-mic colour gradient to remain constant throughout themeasured range of radii: it does not, even for a “purespheroid”, if one is able to extend the measurements, ei-ther near the centre or outside the central body of thegalaxy. There are also significant differences between the


Fig. 5. Experiment showing pseudo-colour profiles induced bydifferential seeing, and their approximate correction. Abscissae:log of isophotal mean radius r in arcsec. Ordinates: Colour inmagnitude, with circles for the majA and crosses for the minA.The model galaxy is the flat object considered in Fig. 4. Thesharper PSF is the same as in the previous figures, but nowthe broader PSF DWY2 has extended wings, elongated in thedirection of the minA of the galaxy. FWHM ’s: 0.67 arcsecalong PSF minA; 0.82 arcsec along PSF majA. Upper curve:Colour profiles for convolved frames of the model. Intermediatecurve: Colour profiles after PSF matching with the best circu-lar Gaussian. Lower curve: Colour profiles after PSF matchingwith the best elongated Gaussian. The correction whith a cir-cular Gaussian is now unsufficient, but it is successful with aproperly elongated Gaussian

gradients in the disk and spheroid of S0’s (and diE?),which translate into differences between the major andminor axis gradients for inclined objects. The presence ofdust leads to various appearances: local features can some-times be avoided in order to get more significant colourgradients. Dust layers in disks lead to minor axis asym-metries in light and colours (MS93), while diffuse dust inspheroids will modify apparent light and colour gradients(see the calculations by Witt et al. 1992). It is thereforeuseful to provide colour data at the relevant level of de-tails, the ultimate being quantitative 2D colour maps.

Fig. 6. Experimental correction of the colour profile of a realgalaxy, i.e. NGC 3377 for the artefacts of differential seeing.Here are considered two frames taken in the same R pass-band with measured PSF’s of 0.59 and 0.66 arcsec FWHM .Abscissae: log of isophotal mean radius r in arcsec. Ordinates:Colour in magnitude, with circles for the majA and crosses forthe minA. The uncorrected R−R colour (upper graph) showsa central red peak and minA blueing as for the model of Fig. 4.After correction (lower graph) the R − R colour becomes flatexcept for noise fluctuations

3.2.1. 1D colour measurements

To provide a good insight into the colour properties ofa given object without necessarily ressorting to the 2Dcolour maps, we have chosen to measure both the radialand azimuthal colour distributions. We use ad hoc com-puter programmes, involving as input the two frames tobe compared and the table of isophotal contours for oneof these: the set of contours is used to locate correspond-ing points in the two frames. Note that for this particularpurpose “symmetrized” contours are used, retaining onlythe even cosine harmonics of their representation.

To get the radial distributions, local colours are aver-aged along arcs of the tabulated isophotes, of length 45degrees in the eccentric anomaly ω of Carter’s ReferenceEllipse. Both major and minor axes, and both halves ofthe two axes, are measured separately. Note that, sincethe isophotes and isochromes are very much alike in E-S0galaxies, averaging the colours along moderate intervals inω of the isophotal contours, is a technically justified wayto improve the S/N ratio.

To get the azimuthal distribution, local colours are av-eraged inside two concentric isophotes sufficiently apart toimprove the S/N ratio, while the azimuthal resolution iskept to 120 points, or 3 degrees in ω. Note that evenly


Fig. 7. Experimental correction of the colour profile of a realgalaxy, i.e. NGC 3115 for the artefacts of differential seeing.Here are considered two frames taken in the same R pass-band with measured PSF’s of 0.55 and 0.91 arcsec FWHM .Abscissae: log of isophotal mean radius r in arcsec. Ordinates:Colour in magnitude, with circles for the majA and crossesfor the minA. Due to widely different PSF’s, the uncorrectedR − R colour (upper graph) shows a strong central red peakand minA blueing as for the model of Fig. 4. After correction(lower graph) the R − R colour becomes much flatter, exceptfor a local slight bump on the minA

distant points in ω are not so in position angle from thegalaxian centre: they are indeed more closely packed nearthe tips of the major axis. Suitable software is also avail-able to get the azimuthal colour distribution as a functionof the PA itself.

Sample outputs of the above measurements are pre-sented in Figs. 1 and 2 for NGC 4125, a galaxy with astrong dust pattern and an exceptionally large colour gra-dient. Then the radial B − R profiles are quite differentfor the 4 mesured arcs, although averaging in rather largeazimuthal domains reduces such differences. For the az-imuthal profiles the averaging is performed in radial do-mains as noted in the figure labels, and more detail of thetrue B−R distributions are kept. A 2D map of the B−Rdistribution for the sample galaxy NGC 4125 is shown inFig. A14 and should be compared with the 1D graphs:it preserves of course more details, but the 1D graphs arecertainly useful to appreciate the significance against noiseof some features of the distributions.

3.2.2. 2D colour maps

The above 1D graphs will indicate if the colour distribu-tion follows the SuBr distribution or presents significant

Fig. 8. Corrections to the radial colour distributions forNGC 4473. The frames had seeing FWHM of 1.19 arcsec inB and 1.05 in R, poorly measured on a faint star. Abscissae:log of isophotal mean radius r in arcsec. Ordinates: Colourin magnitude, with circles for the majA and crosses forthe minA. Upper graph: uncorrected results. Intermediate:adopted correction with a Gaussian of σ = 0.2 arcsec. Lowergraph: “overcorrection” with σ = 0.37. The central red peakprogressively turns out into a blue feature

deviations, due to dust or population variations. In thiscase it is interesting to consider 2D colour maps. We havechosen to produce colour maps strictly consistent with theclassic astrophysical definition of the colour as a differ-ence of magnitudes. This requires a precise alignment ofthe two frames, which is not always easily achieved whenthere are no suitable sets of stars. Very often, one can onlyput in coincidence the centres of the studied galaxy in thetwo images... but this centre may be influenced by colourfeatures (and of course by noise).

A number of techniques have been experimented to re-duce the noise in colour maps, or more exactly to increasethe range of SuBr where this noise remains acceptable,including the one introduced long ago by Sparks et al.(1985): they replace the redder of the pair of frames bya synthetic image built from a set of isophotal contours.It was concluded however that the improvement was notworth the extra work, since the frames studied here areuseful only to measure the innermost range of galaxiancolours.


3.2.3. 1D and 2D asymmetries

Besides 2D colour maps, we have also considered in somecases, 1D graphs of SuBr asymmetries as in MS93, andalso 2D maps of asymmetries, obtained by comparing agiven image with a model of the same, calculated fromthe corresponding file of isophotal contours: one may usea model with elliptical isophotes, as in van Dokkum &Franx (1995), or keep the even cosine Carter’s coefficientse4, e6, ... The model then preserves the diskyness, or boxy-ness, of the true contours. To get asymmetry maps, wealso impose a unique centre and constant orientation tothe model isophotes. Such maps have been produced insuch cases where the asymmetries were thought to bringuseful hints about the dust distribution, in complement tocolour maps. By analogy with the findings of MS93, forS0’s, it was supposed that dust concentrated in the disk ofdiE’s, might give asymmetries in their inner bulge light:this is indeed the case for a few diE’s.

Remark: It should be realized that fitting ellipsesproduces isophotes that are “interlaced” with the realones, so that differences between the true image and themodel necessarily compensate “dark residuals” (due to ex-tinction?), by nearby “bright residuals”. Conversely, thebright residuals due to an embedded disk are compensatedby dark residuals that should not be mistaken for absorp-tion markings. In order to get reliable “extinction maps”,rather elaborate procedures are necessary, as described byGoudfrooij et al. (1994c).

3.3. Calibrations and colour corrections

The frames have been calibrated mostly from the UBV RIaperture photometry by Poulain (1988), and Poulain &Nieto (1994) where the R band is in Cousins’s system.These data are available for 30 objects in the sample. Forthe others, the catalogues of UBV aperture photometryby Longo & De Vaucouleurs (1983, 1985) were used. The Rphotometry was obtained from a tight correlation betweenthe observedB−V and V−R from Poulain (1988). For twogalaxies, i.e. NGC 3613 and 4649, the available field wastoo small for calibration with existing aperture photome-try: the first was calibrated by fitting to calibrated wide-field frames from the Observatoire de Haute-Provence; thesecond by fitting to the B and R data from Peletier et al.(1990) or PDIDC. As regards the Pic du Midi subset of6 objects, 3 were calibrated from Poulain’s photometry.No significant difference with the CFHT bulk of data isexpected.

Poulain’s aperture photometry is accurate enough toprovide good tests of the sky background. If the sky canbe measured with sufficient accuracy from a corner of theframe, the derived photometric zeropoint does not showsystematic changes in the various apertures. If not, as it

was of course the case for large galaxies, the assumed back-ground value was varied until the test yielded satisfactoryresults.

The B−R colours collected in Table 6 have been cor-rected for Galactic reddening and K-effect. It is to benoted that graphical data are not corrected. The B − Vcolour excesses were derived from the B extinction valuesgiven in the Third Reference Catalog of Bright Galaxies,or RC3, by de Vaucouleurs et al. (1991). From Riekeand Lebofsky (1985), we found the galactic colour excessin B − R to be 1.75 larger than E(B − V ). For the Kcorrection the RC3 precepts were followed. and the radialvelocities taken from this same source. The proper coef-ficient was taken from Frei & Gunn (1994). The work ofFukugita et al. (1995) was also considered.

The resulting CFHT colour system has been comparedwith the one of PDIDC, using the observed colours atthe r = 10 arcsec contours for 17 objects in common.Note that PDIDC calibrated 30 of their 39 galaxies fromBurstein et al. (1987) photometry. A zero point differencefor PDIDC-Us of 0.08 is found, with σ = 0.04.

Remark: In the comparison with PDIDC, the galaxyNGC 2768 was neglected: it is found much bluer byPDIDC than here, at 4σ of the above mean difference.

3.4. Errors due to“differential seeing” and theircorrection

By differential seeing, we mean the fact that colour mea-surements involve two frames obtained with different see-ing. As the usually encountered PSF’s have terribly largeeffects upon central SuBr distributions, the difference ofthe two PSF’s will lead to large errors in colours. Theseerrors have been discussed by Vigroux et al. (1988), Franxet al. (1989), Peletier et al. (1990). The later authors de-rived a cutoff radius for each observation and discardedcolours measured inside this limit: in nearly all cases thiscutoff is larger than 3 arcsec, although the conditionsadopted in its definition cannot be considered as verystringent.

Since we are interested in inner colour distributions,i.e. well inside the cutoff radii of PDIDC, we tried to getapproximate corrections for the effects of differential see-ing. The essential step of the corrections is to find a func-tion FC, such as the convolution of the sharper of the twoPSF’s with FC will reproduce the other one. After convo-lution of the sharper frame with FC the colour distributionwill be obtained with the resolution allowed by the worseof the two frames! Another possibility would be to decon-volve the more blurred of the frames by FC, in order tomatch it with the sharper one. Deconvolution artefactsof the kind described by Michard (1996), might be smallenough in this case, because FC is expected to be muchnarrower than the actual PSF. Of these two possibilitiesthe first one has been used in practice, because it is timesaving.


The derivation of FC is of course dependent upon theavailability of a “good” star in the two frames. One canconsider several cases, depending upon the S/N of thestellar images and their actual geometry.

1. If the two PSF’s can be approximated by Gaussianswith circular symmetry, FC is simply another suchGaussian (case C1GR).

2. If the PSF’s have important wings, still with circularsymmetry, FC can be better approximated by the sumof two Gaussians, intended respectively to match thecore and wings of the PSF’s (case C2GR).

3. If the PSF’s are elongated, often due to unequal guid-ing errors in α and δ, FC can be better approximatedby a Gaussian with different σ values in x and y (caseC1GXY). Eventually the sum of two Gaussians mightbe considered (case C2GXY).

These various cases can be implemented by ad hoc MIDASprocedures. Very often however the S/N of available stel-lar images in our small field frames does not allow muchrefinements in the derivation of FC, and one has to besatisfied to use the C1GR approximation.

Several experiments have been made to ascertain theeffects of differential seeing and the success of the abovecorrection techniques. Part of such experiments were madewith model galaxies and model PSF’s. The model ob-ject, assumed colourless, was convolved with two differentPSF’s, a sharper one P1 and a poorer P2. The correspond-ing colour profiles C1 − C2 were evaluated, and then theabove corrections techniques applied. Of course, P1 andP2 are not simple Gaussians, since in this case an exactcorrection is readily obtained. Figures 3 to 5 present theresults of three such experiments.

Other tests were made on real galaxies which have beenobserved twice in different nights and seeing conditions.Then it is possible to measure the pseudo-colours B − Band R − R resulting from the corresponding frames: thisillustrates the effects of differential seeing, and the suc-cess of its correction in actual observations, that is inthe presence of noise. The Figs. 6 and 7 give examplesof these tests, which were applied to 12 pairs of frames.Finally, Fig. 8 shows how the central colour profiles willvary with slight changes in the adopted PSF matchingfunction FC (case C1GR). The central red peak in the“observed” colour of the test object NGC 4473 is lessenedby the adopted correction, i.e. with a correcting Gaussianof σ = 0.2 arcsec. It becomes a blue feature with σ = 0.4.

From the tests here described we draw the followingconclusions:

1. the errors in peak core colours due to differential see-ing may reach several tenths of magnitude, also at therelatively good seeing conditions of the CFHT.

2. such errors, at the level of 0.02 mag, extend only up toa radius of twice the FWHM of the PSF (the worseone) for a roundish object and circular PSF’s. But theeffects are much worse for an elongated PSF “crossing”

the minA of a flattened galaxy (see Fig. 5). In this casethe geometry of the inner isochromes may be seriouslymodified.

3. the errors here discussed may be much reduced bymatching the PSF’s, as described above. The improve-ment is limited by difficulties in getting well definedPSF’s from the noisy images of faint stars.

3.5. Comparison of results for multiply observed objects

For several of the sample galaxies, more than one frameof suitable S/N ratio are available in one or both colours(neglecting very short core exposures!). Such multiple ob-servations may be used to evaluate part of the errors in-volved in the present work. Two different approaches werefound useful.

3.5.1. Pseudo-colours from pairs of frames with the samefilter

The derivation of pseudo-colours B − B or R − R givesuseful information about errors of various origins. The fol-lowing cases should be distinguished in these experiments:

1. For frames where the galaxy is located at widely dif-ferent positions within the instrumental field, or takenwith different instrumentation, the errors in back-ground level or flat-field trends will be uncorrelated.The calibrations may also differ, if the number of aper-ture photometry results is not the same for the twoframes. It was found that large residuals may occurunder these circumstances. It is therefore advisable toderive colours from pairs of frames taken in succes-sion during the same night and without large offsetsof the object within the field. This was the usual prac-tice for the observers who collected the presently usedmaterial.

2. For frames taken with the same instrument and withthe galaxy at nearly the same location on the CCDtarget, the residuals in pseudo-colours are due essen-tially to inaccuracies in the PSF matching. Other er-rors, such as resulting from the background estimateor residual trends in flat-fielding, will be correlated inthe treatment of such parent frames. Results for thesecases have been considered above (see Figs. 6 and 7).

3.5.2. Comparisons of colour distributions from differentpairs of frames

Such comparisons could be achieved for 8 galaxies: twowere observed on two consecutive nights, the second withbetter seeing; four were re-observed with the HRCam inthe hope to get better resolution. For the other there wassome duplication of the data in a single night, a case of


Table 5. Comparison of B−R data from multiple observations.Fr: code for the pair of frames, for reference to Table 1. WB:FWHM in B. WR: FWHM in R. C0: B − R at core centre.C1: B − R at radius r = 1 arcsec. C3: B − R at radius r = 3arcsec. Gr: logarithmic B − R gradient for r > 3 arcsec. Thecolours are here uncorrected. Nt: Notes to Table 3 (a) No star;mean FWHM ’s for the night. (b) Galaxy near the edge of theframe to get a star in. (c) HRCam observations

NGC Fr WB WR C0 C1 C3 Gr Nt3377 h05 1.0? 0.9? 1.57 1.49 1.46 −0.07 (a)id h06 0.77 0.58 1.57 1.53 1.48 −0.12 (b)4406 h04 0.67 0.74 1.60 1.58 1.56 −0.02 -id h20 0.85 0.59 1.64 1.63 1.60 −0.07 (c)4472 h04 0.77 0.72 1.67 1.66 1.64 −0.06 -id h20 0.82 0.72 1.65 1.64 1.61 −0.10 (c)4478 h04 0.86 0.72 1.57 1.50 1.46 −0.05 -id h07 0.62 0.50 1.54 1.49 1.45 −0.02 (c)4494 h04 1.2? 0.75 1.62 1.56 1.44 −0.11 -id h20 0.89 0.79 1.66 1.57 1.48 −0.07 (c)

limited interest because the differences are due mainly toerrors in PSF matching, already discussed above. Here arediscussed only the cases where the errors in the B−R dis-tributions are largely independent, except for the errors ofcalibration. Table 5 presents the results of these compar-isons, using ad hoc parameters.

The results of the present experiments are summarizedbelow, both from pseudo-colours and from multiple colourobservations:

1. Errors due to imperfect PSF matching are restricted toa radius roughly equal to the worse PSF FWHM . Anestimate of random errors upon the central core colouris 0.03. For strongly flattened objects errors upon theminA colour profile may occur at the same amplitude(see the case of NGC 3115, in Fig. 4).

2. Spurious colour patches at an amplitude of 0.03 mayoccur due to unsufficient S/N .

3. Rather large errors may develop near the limits of theavailable field. This is due to poor background esti-mates: these may be worse than in classical observa-tions, where the sky light is effectively registered onthe frame due to adequate field of view and gener-ous exposure. As a result the logarithmic gradients ofB − R are quite uncertain. The experiment summa-rized in Table 5 point to a mean difference of 0.05 be-tween two measures of the gradient for the same object!Similarly the mean difference between two measures ofthe colour at re/2.5 is 0.04. The mean errors upon asingle measurement will be slightly smaller.

3.6. Classification of central SuBr profiles

It has been shown by Nieto et al. (1991a), and more re-cently by Jaffe et al. (1994) and Lauer et al. (1995), thatthe “cores” of E-galaxies can be sorted out in two types,here termed flat topped core and sharp peak, or respectivelyftc and shp. This corresponds to Type I and Type II in

the notation of Jaffe et al., or “core-like profile” against“power-law profile” in Lauer et al. wording.

In view of a comparison of central colour profiles withthe types of central SuBr profiles, it was necessary tosupplement the lists of “core” types available from thequoted papers. For this purpose, the R frames were de-convolved by Lucy’s technique, as implemented in theMIDAS software, using 27 iterations. Then three param-eters were examined: change of peak SuBr between theoriginal and deconvolved frames, or equivalently the ratioof the FWHM ’s in the original and deconvolved frames,and finally the FWHM of the deconvolved core. Thesethree parameters indeed show a bimodal distribution. Forthe first two, this corresponds to the fact that ftc pro-files are resolved, or nearly so, at the CFHT resolution,while the shp profiles remain quite unresolved. The lastone is less dependent upon the actual PSF: it would per-haps converge towards an exact galaxian property if thenumber of iterations was varied in relation with the frameresolution... and if the PSF were perfectly accurate.

The present classification of central profiles, given inTable 6, shows perfect agreement with the one of Faberet al. (1997) for 15 galaxies in common. From the graphsof Byun et al. (1996), there are a few profiles interme-diate between the typical “core-like” and “power-law”cases. Similarly our classification gives uncertain resultsfor NGC 4125, 5322, 5576, 5638. Not that the few objectsobserved at the TBL could not be classified.

4. Remarkable features in the B −R distributions inE-galaxies

According to the present study, the colour distributions inE-galaxies may feature the following properties:

1. Colour patterns are made up of localized red patches,and of straight or curved lanes, sometimes forminga complex system. Such local features are unambigu-ously assigned to dust.

2. The dust in disk phenomenon corresponds to a set ofproperties in the colour distributions of galaxies con-taining a disk, and denoting a concentration of dust inthis component. The most obvious of such propertiesis a redder lane along the disk or close to it.

3. The central red peak stands out in many colour dis-tributions, even after careful matching of the PSF’s,as described in Sect. 3. Arguments will be given sug-gesting that a central dust concentration might be thecause of this feature in many cases, but no conclusiveevidence is available.

4. Blue features are scarce enough to be considered asanomalous in E-galaxies.

It should be emphasized that the above listed features arenot exclusive of one another, and can very well occur to-gether in a single object.


Fig. 9. Sample of objects with flat radial B − R distributions.The label of each graph is the name of the plotted file, whichidentifies the galaxy. Abscissae: log of isophotal mean radius rin arcsec. Ordinates: Colour in magnitude. We note that suchobjects are large galaxies with flat topped cores

4.1. Local red “colour patterns”

Local dust features are detected in 16 galaxies out of 42in our sample, leaving aside the two anomalous blue ob-jects NGC 3156 and 4742. These formations differ widelyin size and contrast, and it was deemed useful, for classi-fication purposes, to introduce a dust pattern importanceindex (DPII), in a scale of 0 to 3. The various steps of theDPII can be defined as follows:

. 0: No evidence of local dust in B −R data,

. 1-: Some evidence for local dust is detected,

. 1: Clear dust pattern around core,

. 2: Marked dust pattern around core,

. 3: Outstanding dust pattern.For the indices 2 and 3, the dust produces visible distor-tions in the B isophotes.

In Paper I of this series, this index was derived for 67 Egalaxies of the Local Supercluster, both from the presentsample and from literature data (including many multiple

Fig. 10. Sample of objects with radial B − R distributions in-volving a central isolated red peak. The label of each graphis the name of the plotted file, which identifies the galaxy.Abscissae: log of isophotal mean radius r in arcsec. Ordinates:Colour in magnitude. We note that such objects also have asharply peaked central light distribution

observations), and a statistical discussion was given, relat-ing the DPII to the morphological subclasses, diE, unEand boE, and to other suggestive groupings of objects.

The colour patterns in E’s are centrally located, andthe galaxian centre is usually the reddest point in the dis-tribution. The pattern of NGC 2974 is an exception, withthe peak colour displaced by some 1.5 arcsec.

The strong dust patterns in E galaxies show a geat va-riety of mostly irregular geometries. Regular rings, as seenin S0 and Sa galaxies at various inclinations are scarcelyencountered. The complex colour pattern of NGC 2974shows however several such rings.

It is perhaps useful to bring attention to a number ofdust patterns of special interest as displayed in the maps ofFigs. A1 to A32. Outstanding dust patterns (DPII = 3)occur in NGC 1052, 2768, 2974, 4125 and 4374. The caseDPII = 2 is represented only by NGC 5831. Is it a subjec-tive trend of our selection or the result of a gap in the real


Fig. 11. Radial distribution of B−R for NGC 3156. Abscissae:log of isophotal mean radius r in arcsec. Ordinates:Uncorrected colour in magnitude. Circles: majA; crosses:minA. This anomalous galaxy contains a centrally locatedvery blue region and also some dust

Fig. 12. Radial distribution of B−R for NGC 4742. Abscissae:log of isophotal mean radius r in arcsec. Ordinates:Uncorrected colour in magnitude. Circles: majA; crosses:minA. This anomalous galaxy contains a centrally locatedvery blue region with a red peak at the very centre. Theavailable field was quite small so that the colour farther thansome 8 arcsec is very uncertain

distribution? The classification of a much larger samplewould be necessary to decide.

For DPII = 1, we note, among others, NGC 0821,with a short dust lane nearly along the minor axis; andNGC 4660 with a dust lane nearly along the major axis,so that it is also classified among the objects with “dustin disk”.

The DPII value 1- has been given in cases wherethe isochromes appear to deviate significantly from theisophotes, without a well contrasted colour pattern.Examples are NGC 3377 and 3379. Another case ofinterest is NGC 4261, where dust is not clearly seen in

the B−R distribution, but appears as a small absorptionarclet in the B image. This minute feature as beenregistered with the HST (see Jaffe et al. 1994). There aredoubtful cases, where the 1- index could become 0 witha better S/N ratio, or conversely. For NGC 4494 we findDPII = 0 (no dust), while Goudfrooij et al. (1994b) andvan Dokkum & Franx (1995) see dust near the nucleus ofthis object.

As noted in Paper I, there is generally no evidencefor dust patterns in boE galaxies. We refer to the mapsof NGC 4365, 4406, 4472 for examples of giant dustlessboE’s, and to NGC 4387 and 4478 for minor boE’s withrotational support (see Nieto & Bender 1989). The lat-ter have no colour pattern but a sharp red nucleus (seebelow).

4.2. “Dust in disk” phenomenon

In Paper II of this series, the evidence has been consideredfor dust concentration near the disk of E galaxies con-taining a disk component. This involved the disk of diE’sat sufficient inclinations, but also the inner “decoupled”disks seen in both boE’s and diE’s. This evidence mayinclude the “minor axis asymmetry” in light and colour,described and modelled for S0 galaxies in MS93. A distinctreddening of the major axis in the region where the diskcontributes significantly to the SuBr might be also takenas evidence for “dust in disk”, but is somewhat ambigu-ous, because a similar appearance may result from verydifferent colour gradients between the spheroid and diskof a two-component galaxy.

Looking at the set of maps below, the two above prop-erties appear as a flattening of the isochromes as comparedto the isophotes, and eventually their displacement in theminor axis direction, relative to the centre of symmetry ofthe isophotes. These appearances can be recognized in thegraphs pertaining to NGC 0584, 3585, 3610, 4473, 4621,5322, 5845. The latter two objects have inner decoupleddisks, and their HST deconvolved images appear in Laueret al. (1995).

In the summary data table of Sect. 5, galaxies showingevidence of the “dust in disk” phenomenon are noted withthe dd symbol.

4.3. Red nuclei

Red nuclei in E-galaxies have been considered by Sparkset al. (1985): they note “an unresolved nucleus 0.3 magni-tudes redder than the surrounding galaxy” for NGC 6958.This description obviously refers to an artefact of differ-ential seeing, according to the experiments described inSect. 3, and to previous studies referenced there. In thepresent work, the PSF’s of the two frames involved in acolour measurement were equalized, and the red nuclei


can be objectively studied. The resolution is limited bythe worse of the two frames, the B one in general, andthe accuracy is estimated to 0.02− 0.03 magnitude at thecentre and nearby regions.

The central regions show the following types of colourdistributions:

1. nearly flat. This is the case for a few large galaxies, i.e.NGC 0720, 4365, 4406, 4472, 4649, 5846 and perhapsalso for the compact dwarf NGC 4486B. Such cases areillustrated in Fig. 9. In this situation, the logarithmiccolour gradient tends to decrease inwards of a limitingradius of a few arcsec (much less for the dwarf). Wenote from Table 6 that such objects are in the “flattopped core” (ftc) class of central SuBr profiles;

2. isolated red peaks. In this case, illustrated in Fig. 10,the logarithmic colour gradient increases inwards ofa break radius rb, with typical values 1.5 < rb < 3arcsec, for the range of distances and object sizesof our sample. The amplitude of the red nucleus isparametrized by the difference ∆C0,3 of the B − Rcolours at the centre and at the radius r = 3 arcsec.Objects with isolated red peaks of large ∆C0,3, notablythe examples of Fig. 10, usually belong to the “sharppeak” (shp) class of central SuBr sprofiles;

3. embedded red peaks. In this case, the red nucleus ispart of a larger colour pattern, and the radius rb isno more defined. The colour difference ∆C0,3 can stillbe measured but cannot be looked at as a property ofthe red nucleus itself. The 1D colour graphs are not ofmuch use in such cases. One example has been given inFigs. 1, 2 for NGC 4125.

It is tempting to speculate that central red peaks may becaused by a concentration of dust near the centre of theobject (although population variations might also be partlyresponsible?). This is probably the case for red peaks em-bedded in a larger dust pattern. Since minute central dustpatterns have been discovered from HST frames one mightconsider these as a likely cause of the red peaks seen at alesser resolution.

Remark: There are two atypical cases deviating fromthe above classification of red nuclei. NGC 0636 exhibitsa small blue nucleus at the centre of a normal red nucleus;for NGC 2974, the peak B−R is displaced some 1.5 arcsecfrom the galaxian centre.

4.4. Anomalous objects

In our sample, there are 2 galaxies with a central regionmuch bluer than its surroundings, most probably due torecent star formation.

NGC 3156 is an E5 in the RSA, but has type T = −2in the RC3. From the same catalogue, its colours are veryblue, 0.75 in (B − V )e and 0.29 in (U −B)e. The radialB − R distribution is shown in Fig. 11. This galaxy alsoshows a small dust arclet.

NGC 4742 is an E4 in the RSA, and has type T = −5in the RC3. The catalogue colours are 0.80 for (B − V )T

and 0.30 for (U −B)T. Colours in the effective radius arenot available. The radial B − R distribution is shown inFig. 12. A sharp red peak occurs at the centre of theobject.

5. Data and correlations

5.1. Data description

The B − R data derived from the Nieto’s sample is pre-sented as follows:

1. Table 6 lists global characteristics of the studied galax-ies and selected numerical, or coded, colour data. Notethat reference isophotal colours in the Table were mea-sured outside dust patterns, as well as feasible, andcorrected for galactic extinction and K-effect as indi-cated above, the adopted corrections in B − R beinggiven in the table.

2. The set of Figs. A1 to A32 shows maps of the B − Risochromes and B isophotes in selected small fields, ofgenerally adequate S/N ratio, near the centre of eachobject.

3. Maps of B − R in FITS format are available inelectronic form, upon request to the author. A fewsamples are available in the anonymous space of theNice serveur: access by ftp ftp.obs-nice.fr; directoryusers/bigfic/anonymous/pub/michard.

5.2. Comparison with other surveys

Reference colours Previous surveys by Franx et al. (1989)(FIH), Peletier et al. (1990) (PDIDC) and Goudfrooijet al. (1994) (Gea) contain reference colours measured atthe isophotal contour of radius re/2. We give in Table 6 aB−R isophotal colour C2 measured at re/2.5, more easilyreached in our small field frames. Attempts to plot colour-colour graphs from these data gave somewhat surprisingresults, even using the two colours from the same survey.It has been supposed that some sets of data involve muchlarger errors than indicated by the authors. To get someinsight into this question, we compared the various setsof reference colours to the mean colours inside re derivedfrom aperture photometry by Poulain (1988) or Poulain& Nieto (1994), supplemented in a few cases by our ownestimates from the interpolation of available aperture pho-tometry. Obviously, the isophotal colours at re/2 and themean colours within re are not expected to be strictlyequal, but cannot differ very much: the former might besomewhat bluer because the latter are possibly influencedby the central dust patterns. For these comparisons theobserved mean colours were corrected using the combinedcorrections applied in each of the surveys.


The results of the described tests are as follows:. (B −R)FIH − (B −R)0

e :N = 17; Mean = 0.276; σ = 0.008. (U −B)FIH − (U −B)0

e :N = 14; Mean = −0.030; σ = 0.042. (B −R)PDIDC − (B −R)0

e :N = 24; Mean = 0.041; σ = 0.050. (U −B)PDIDC − (U −B)0

e :N = 24; Mean = −0.009; σ = 0.079. (B − V )Gea − (B − V )0

e :N = 46; Mean = −0.000; σ = 0.036. (V − I)Gea − (V − I)0

e :N = 39; Mean = −0.017; σ = 0.166. (B −R)Us − (B −R)0

e :N = 41; Mean = −0.006; σ = 0.039.

The following suggestions may be made from the nu-merical results, supplemented by graphs of the comparedquantities:

1. The B−R colours of FIH are not in Cousins’s systemas stated in their paper. A constant correction is ade-quate to bring their data into this system. As regardstheir U −B colours, they generally agree well with the(U − B)0

e . In view of the small number of objects, 3values too blue by 0.1 are enough to explain the meanand σ above.

2. The B − R colours of PDIDC are somewhat redderthan indicated by the here used aperture photometry.This was also found in a direct comparison with ourresults. Their U−B colours contain a few very unlikelyvalues: NGC 2768 is found much too red, NGC 4486and 5638 much too blue (at more than 2σ).

3. The B−V colours of Gea are in good agreement withthe (B−V )0

e values. On the other hand their V −I dataare plagued with a number of “impossible” values, ei-ther too red (i.e. NGC 720) or too blue (i.e. NGC 3377,4564, 5813) by more than 2σ, that is more than 0.33.

4. Our B − R at re/2.5 are in good agreement with the(B −R)0

e , as expected from our calibration sources.

Colour gradients Although the present material is farfrom ideal, as emphasized before, to measure the smallcolour gradients in E-type galaxies, we have looked atthe correlations between our “outer gradient” G12 (seeTable 6) and the results from other surveys. We have alsocompared the gradients for the two available colours inthe PDIDC and Gea surveys. This was done by calculat-ing the impartial correlation between centred variables,eventually rejecting extremely divergent values. The re-sults are as follows:

1. If x = −∆(B − V )Gea/∆ log r and y = −∆(B −I)Gea/∆ log r we find for 42 objects (2 rejected) y =1.95x+0.009 with a coefficient of correlation ρ = 0.75.Thus the gradients in two colours from these authorsare mutually consistent.

2. If x = −∆(B − R)PDIDC/∆ log r and y = −∆(U −R)PDIDc/∆ log r we find for 37 objects (2 rejected)y = 2.20x + 0.036 with a coefficient of correlationρ = 0.51. The correlation between the gradients ofPDIDC in the two colours appear “weaker than mightbe expected” to quote these authors. Note that thetwo B−R and U−R gradients of FIH are very poorlycorrelated (with only 14 data points).

3. If x = −∆(B − V )Gea/∆ log r and y = −∆(B −R)Us/∆ log r we find for 21 objects y = 1.40x− 0.019with a coefficient of correlation ρ = 0.60. Our B − Rgradients are reasonably consistent with the B − Vgradients of Gea.

4. If x = −∆(B − I)Gea/∆ log r and y = −∆(B −R)Us/∆ log r we find for 20 objects y = 0.73x− 0.027with a coefficient of correlation ρ = 0.82. Our B − Rgradients are again reasonably consistent with theB − I gradients of Gea.

5. If x = −∆(B − R)PDIDC/∆ log r and y = −∆(B −R)Us/∆ log r we find for 19 objects y = 0.92x+ 0.013with a coefficient of correlation ρ = 0.41. Our B − Rgradients do not correlate as well with the gradients ofPDIDC as with those of Gea.

Dust patterns The mappings of large dust patterns by var-ious authors are generally in reasonable ageement. This isthe case when comparing our data with the V − I mapof NGC 1052 by Sparks et al. (1985), or the maps ofGoudfrooij et al. (1994b), for NGC 2974, 3377, 4125, 4374,4660.

Discrepancies arise, when one of the intercomparedsurveys is of much lesser resolution, or perhaps affectedby the effects here described as “differential seeing”.For instance, Goudfrooij et al. find a blue inner disk inNGC 3610 and 4473, in contradiction with our data.An interesting case is NGC 4494: for this object V − Imaps have been obtained from HST frames by Forbeset al. (1995), and by Carollo et al. (1997). These showa dust ring of subarcsec radius, obviously located inthe inner disk, and much stronger on the W side of themajor axis. We have overlooked this feature, although itproduces in our B − R map (Fig. A23) a clear E − Wcolour asymmetry (for this same object, Goudfrooij et al.note a minor axis dust lane). NGC 4494 is an idealexample in favour of our speculation above, that a centralaccumulation of dust might be responsible for the isolatedred nuclei found in a number of objects of our shp class.From a glance at the V − I radial profiles and V − Iimages of the above quoted authors, it seems that severalother galaxies of the “power-law” class have both nucleardust and a sharp red peak.

Dust features may well be missed in the present surveydue the poor S/N ratio of part of our frames (quoted P inTable 1). For instance a faint dust pattern is found near


Table 6. Isophotal colours at selected radii and gradients. MT: Absolute magnitude in B from Michard & Marchal (1994), orderived accordingly. logre: Logarithm of the effective isophotal radius in arcsec from same source. ∆(B − R); Corrections tothe observed B −R for galactic extinction and K-effect. CP: ftc for a flat topped core, shp for a sharp peak. C0: Peak centralcorrected B − R colour. ∆C0,3 colour difference between centre and radius r = 3 arcsec. C1: Corrected B − R at re/10. It isuncertain, or not measured, if re < 10. C2: Corrected B−R at re/2.5. It involves an extrapolation for large objects. G12: Outerlogarithmic gradient for r > 3 arcsec. Code: Dust pattern importance index DPII and code dd for “dust in disk”. A colon:refers to an uncertain result.

Notes to Table 6: (a) NGC 0636: small blue dot in core: representative core colour interpolated. (b) NGC 2974: peak B−Rin patch 1.7 arcsec NW of centre. (c) NGC 3156: Very anomalous colour distribution (see Fig. 11) (d) NGC 4125: Extraordinarygradient (see also Goudfrooij et al. 1994a) (e) NGC 4742: Very anomalous colour distribution (see Fig. 12) (f) NGC 5322: Innersmall disk, with some reddening, isolated from main boxy body. (g) NGC 5845: Inner small disk, with some reddening, isolatedfrom main disky body

NGC Type −MT logre ∆(B −R) CP C0 ∆C0,3 C1 C2 G12 Code -

0584 diE 19.82 1.37 0.08 shp 1.57 0.11 1.47 1.43 −.03 0 dd -0596 diEp 19.66 1.43 0.07 shp 1.57: 0.07 1.52 1.47 −.04 1- -0636 diEp 18.98 1.31 0.08 shp 1.52 0.06: 1.48 1.42 −.07 0 (a)0720 unE 20.85 1.53 0.02 ftc 1.62 0.02 1.60 1.56 −.05: 0 -0821 diE 20.54 1.51 0.09 shp 1.63 0.05 1.59 1.54 −.07 1 dd: -1052 diE 19.79 1.37 0.04 shp 1.82 0.19 1.62 1.53 −.10: 3 -2768 diE 21.13 1.70 0.09 shp 1.85 0.16 1.67 1.61 −.11: 3 -2974 diE 20.41 1.38 0.07 shp 1.63 0.09 1.55 1.52 −.11: 3 (b)3156 SA0 17.62 1.17 0.03 - 0.66 - 1.12 1.24 +.20 1- (c)3193 unE 19.68 1.21 0.05 - 1.55 0.07 1.50 1.47 −.05 0 -3377 diE 19.25 1.56 0.04 shp 1.56 0.10 1.46 1,40 −.10 1- dd -3379 unE 20.11 1.75 0.03 ftc 1.62 0.05 1.55 1.52 −.05 1- -3585 diE 20.05 1.30 0.14 shp 1.57 0.06 1.51 1.46 −.06 0 dd -3605 boE 17.99 0.87 0.01 - 1.51 0.14 - 1.38 −.10: 0 -3608 boE 19.63 1.56 0.02 ftc: 1.60 0.08 1.52 1.50 −.03 1- -3610 diE 19.06 1.13 0.02 shp 1.56 0.10 1.48 1.43 −.06 0 dd -3613 diE 18.93 1.33 0.03 shp 1.54 0.06 1.49 1.47 −.03 1-: -3640 boE 19.85 1.34 0.06 - 1.57 0.09 1.48 1.45 −.04 0 -3872 diEp 20.55 1.13 0.09 - 1.55 0.05 - 1.50 −.01: 0 -4125 diE 20.55 1.55 0.03 shp 1.80 0.20 1.60 1.44 −.25 3 (d)4261 boE 20.98 1.59 0.03 ftc 1.68 0.10 1.57 1.54 −.05 1- -4365 boE 20.65 2.00 0.02 ftc 1.61 0.02 1.55 1.53: −.03 0 -4374 unE 21.15 2.02 0.06 ftc 1.79 0.18 1.56 1.53: −.04: 3 -4387 boE 18.90 0.96 0.04 shp 1.60 0.07 - 1.53 −.08 0 -4406 boE 21.44 2.20 0.04 ftc 1.59 0.03 1.52 1.50: −.04 0 -4472 unE 21.65 2.05 0.01 ftc 1.65 0.03 1.57 1.53 −.06 0 -4473 diE 19.92 1.45 0.04 ftc 1.64 0.11 1.53 1.50 −.05 0 dd -4478 boEp 18.78 1.03 0.06 shp 1.53 0.11 - 1.39 −.04 0 -4486B unEp 17.63 0.59 0.06 ftc 1.57 0.04 - 1.53 −.00 0 -4494 unE 19.57 1.56 0.04 shp 1.60 0.15 1.43 1.36 −.09 0: -4551 boE 18.06 1.13 0.08 shp 1.60 0.13 - 1.43 −.06 0 -4564 diE 18.99 1.25 0.05 shp 1.64 0.12 1.56 1.48 −.15 1- dd -4621 diE 20.18 1.68 0.03 shp 1.65 0.10 1.52 1.47: −.08: 0 dd -4649 boE 21.10 1.85 0.03 ftc 1.63 0.00 1.61 1.58 −.04: 0 -4660 diE 18.55 1.09 0.01 shp 1.65 0.09 - 1.51 −.08 1 dd -4742 SA0: 18.15 0.65 0.03 - 1.14 - 1.07 1.00 +.40 0 (e)5322 boE 20.44 1.54 0.03 shp: 1.55 0.13 1.41 1.36 −.09 0 dd (f)5576 boEp 20.16 1.55 0.03 shp 1.50 0.11 1.38 1.34 −.06 0 -5638 unE 19.50 1.35 0.04 shp: 1.56 0.05 1.52 1.48 −.06 0 -5813 unE 20.66 1.59 0.09 ftc 1.62 0.09 1.52 1.47 −.10 1 -5831 diE 19.62 1.33 0.08 shp 1.60 0.11 1.51 1.46 −.10 2 -5845 diE 19.58 0.57 0.08 shp 1.56 0.09 1.49 1.44 −.09 0 dd (g)5846 unE 21.00 1.67 0.08 ftc 1.67 0.03 1.61 1.54: −.06 0 -5846A unE 17.40 0.35 0.08 shp 1.65 0.05 - 1.59: −.04: 0 -


the center of NGC 5846 by Goudfrooij & Trinchieri (1998),which was not detected from our low exposure frames.

6. Concluding remarks

This paper has been devoted to the description of tech-niques and the presentation of data about the B−R colourdistributions in E-type galaxies, as measured from rathergood resolution frames. The scientific results have beengiven in the three previous paper of this series and againsummarized in the Introduction.

Therefore the concluding remarks will be devoted totechnical matters.

1. It has been proven that colour measurements can beextended inside the “seeing” degraded regions of galax-ian images, if a suitable treatment is applied to equal-ize the different PSF’s in the two frames involved inany colour. This treatment may be a convolution ofthe “best” frame by a correcting function, or even-tually the deconvolution of the “worse” by this samefunction. This procedure is successful only if the twoframes have similar “seeing”, so that the correctingfunction is narrow enough, and well measured from asuitable star in the field.

2. The short discussion of available colour surveys ofE-type galaxies in the above section, suggests that thedata are not in a well defined and unique colour sys-tem, and eventually suffer from serious calibration er-rors for a number of objects. A reconsideration of thedata, and perhaps new observations, may be needed toobtain a fully reliable set of colours and their gradients,in the usual optical colours, for a significant sample ofE-type objects.

Appendix A:

This contains maps of the B − R isochromes (full lines)and B isophotes for most of the galaxies in the presentsample, at least if the data was of acceptable S/N ratio.A few maps of poor S/N were included if thought to con-vey some useful information. Since the HRCam has to berotated to get both the object and the guiding star in thefield, the approximate direction of the North is given inthe captions. No image reversal occurs, so that the Eastis always 90◦ counterclockwise from the North.

References

Arnaboldi M., Da Costa G.S., Saha P., 1996, “The Nature ofElliptical Galaxies”, 2d. Stromlo Symp., Canberra, 26-30Aug. 1996, ASP Conf. Ser., 1997

Balcells M., Peletier R.F., 1994, AJ 107, 13Burstein D., Davies R.L., Dressler A., et al., 1987, ApJS 64,

601Byun Y., Grillmair C.J., Faber S.M., et al., 1996, AJ 111, 1889

Carollo C.M., Franx M., Illingworth G.D., et al., 1997, ApJ481, 710

Carter D., 1978, MNRAS 182, 797van Dokkum P.G., Franx M., 1995, AJ 110, 2027Faber S.M., Tremaine S., Ajhar E.A., et al., 1997, AJ 114, 1771Franx M., Illingworth G., Heckman T., 1989, AJ 98, 538 (FIH)Forbes D.A, Franx M., Illingworth G.D., 1995, AJ 109, 1988Frei Z., Gunn J.E., 1994, AJ 108, 1476Fukugita M., Shimasaku K., Ichikawa T., 1995, PASP 107, 945Giudicelli M., Michard R., 1993, Proceedings 5th ESO/ST-

ECF Data Analysis Workshop, p. 105Goudfrooij P., Hansen L., Jorgensen H.E., et al., 1994a, A&AS

104, 179 (Gea)Goudfrooij P., Hansen L., Jorgensen H.E., et al., 1994b, A&AS

105, 341Goudfrooij P., de Jong T., Hansen L., et al., 1994c, MNRAS

271, 833Goudfrooij P., 1995, The Messenger 79, 31Goudfrooij P., 1996, in “The Nature of Elliptical Galaxies”,

2d. Stromlo Symp., Canberra, 26-30 Aug. 1996, ASP Conf.Ser., 1997, Arnaboldi M., Da Costa G.S., Saha P. (eds.)

Goudfrooij P., Trinchieri G., 1998, A&A 330, 123Jaffe W., Ford H.C., O’Connell R.W., et al., 1994, AJ 108,

1567Lauer T.R., Ajhar E.A., Byun Y., et al., 1995, AJ 110, 2622Longo G., de Vaucouleurs A., 1983, A General Catalogue of

Photoelectric Magnitudes and Colours in the UBV Systemof 3578 Galaxies brighter than the 16th V -magnitude,1936-1982, The Univ. of Texas Publ.

Longo G., de Vaucouleurs A., 1985, Supplement to theGeneral Catalogue of Photoelectric Magnitudes andColowith circles for the majA and crosses for the minA. rsin the UBV System, The Univ. of Texas Publ.

Michard R., Simien F., 1993, A&A 274, L25 (MS93)Michard R., 1994, A&A 288, 401Michard R., 1996, A&AS 117, 583Michard R., Marchal J., 1994, A&AS 105, 81Michard R., 1998a, A&A 334, 453Michard R., 1998b, A&A 335, 479Michard R., 1998c, A&A 335, 49Nieto J.L., Bender R., 1989, A&A 215, 266Nieto J.L., Bender R., Arnaud J., et al., 1991a, A&A 244, L25Nieto J.L., Bender R., Surma P., 1991b, A&A 244, L37Peletier R.F., Davies R.L., Illingworth G.D., et al., 1990, AJ

100, 1091 (PDIDC)Peletier R.F., Balcells M., 1996, AJ 111, 2238Poulain P., 1988, A&AS 72, 215Poulain P., Nieto J.L., 1994, A&AS 103, 573Rieke G.H., Lebofsky M.J., 1985, ApJ 288, 618Silva D.R., Elston R., 1994, ApJ 428, 511Silva D.R., Wise M.W., 1996, ApJ 457, L15Silva D.R., Boroson T.A., Thompson I.B., et al., 1989, AJ 98,

131Sparks W.B., Walf J.V., Thorne D.J., et al., 1985, MNRAS

218, 87de Vaucouleurs G., de Vaucouleurs A., Corwin H.G.Jr., et al.,

1991, Third Reference Catalogue of Bright Galaxies.Springer Verlag, New-York

Vigroux L., Souviron J., Lachieze-Rey M., et al., 1988, A&AS73, 1

Wise M.W., Silva D.R., 1996, ApJ 461, 155Witt A., Thronson H.A.jr, Capuano J.M., 1992, ApJ 393, 611


Fig. A1. NGC 0584: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.475−1.60 (0.025). B isophotes: 17.0−19.0(0.5). The flattening of the isochromes, and their shift alongthe minor axis, suggest dust concentration in the disk

Fig. A2. NGC 0596: diEp. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.57−1.62 (0.025). B isophotes: 17.2−19.7(0.5). The slightly tilted, low contrast red pattern, suggestsonly little dust. Possible blue core feature

Fig. A3. NGC 0636: diEp. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.525−1.575 (0.025). B isophotes: 17.0−19.0(0.5). The colour pattern shows a central blue feature, whichseems not to be due to inadequate PSF matching

Fig. A4. NGC 0720: unE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.58−1.595 (0.015). B isophotes: 18.0−19.5(0.5). The colour distribution of this flat core galaxy is also re-markably flat. See also Fig. 9


Fig. A5. NGC 0821: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.675−1.725 (0.025). B isophotes: 17.5−19.5(0.5). A low contrast red lane nearly along the minor axis isthe notable feature of the modest dust pattern

Fig. A6. NGC 1052: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.60−1.80 (0.025). B isophotes: 17.2−19.2(0.5). The object presents a complex dust pattern with a promi-nent lane near the minor axis

Fig. A7. NGC 2768: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.775−1.975 (0.025). B isophotes: 17.5−20.0(0.5). The object presents a complex dust pattern with exten-sive diffuse lanes mostly along the minor axis

Fig. A8. NGC 2974: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.675−1.80 (0.025). B isophotes: 17.0−20.0(0.5). The object presents a complex dust pattern with multi-ple arclets. The peak reddening is well away from the galaxiancenter


Fig. A9. NGC 3377: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.49−1.59 (0.025). B isophotes: 16.5−19.0(0.5). An asymmetric red feature is limited by a discontinuousarclet. See also HST frame in Lauer et al. (1995)

Fig. A10. NGC 3379: unE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.64−1.69 (0.025). B isophotes: 16.5−18.5(0.5). The core red pattern is tilted with a narrow inner lane

Fig. A11. NGC 3585: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.63−1.68 (0.025). B isophotes: 16.6−18.6(0.5). The flattening of the isochromes, and their shift alongthe minor axis, suggest some dust concentration in the innerdisk

Fig. A12. NGC 3610: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.47−1.57 (0.025). B isophotes: 16.75−18.75(0.5). The flattening of the isochromes, and their shift alongthe minor axis, suggest some dust concentration in the innerdisk


Fig. A13. NGC 3613: diE. Positions in ′′. N at 9 o’clock.B−R isochromes: 1.50−1.55 (0.025). B isophotes: 17.0−19.5(0.5). The frames are of poor S/N and small usable field. TheB −R pattern is of low contrast and suggests only little dust

Fig. A14. NGC 4125: diE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.60−1.85 (0.025). B isophotes: 17.5−20.0(0.5). This object shows an important and complex dust pat-tern. Its colour gradient is exceptionally large also farther outthan this pattern. See also Figs. 1 and 2

Fig. A15. NGC 4261: diE. Positions in ′′. N at 2 o’clock.B−R isochromes: 1.65−1.75 (0.025). B isophotes: 16.7−18.75(0.5). A small dust arclet is seen in B light just below thecenter, and greatly disturbs the maximum SuBr contour hereplotted

Fig. A16. NGC 4365: boE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.575−1.60 (0.025). B isophotes: 17.5−19.5(0.5). The colour distribution of this flat core galaxy is also re-markably flat. Compare with Carollo et al. (1997)


Fig. A17. NGC 4374: unE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.70−1.85 (0.025). B isophotes: 17.5−19.5(0.5). The large asymmetric dust pattern is sharply boundednorthwards by a prominent lane

Fig. A18. NGC 4387: boE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.585−1.66 (0.025). B isophotes: 18.2−19.7(0.5). This boE, a member of the family of rotationally sup-ported minor objects, has no significant colour pattern

Fig. A19. NGC 4406: boE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.58−1.595 (0.015). B isophotes: 17.0−19.5(0.5). The colour distribution of this flat core galaxy is also re-markably flat. See also Fig. 9, and the V − I data by Carolloet al. (1997)

Fig. A20. NGC 4472: unE. Positions in ′′. N at 12 o’clock.B −R isochromes: 1.62− 1.63 (0.01). B isophotes: 17.0− 18.5(0.5). The colour distribution of this flat core galaxy is alsoremarkably flat. See also Fig. 9


Fig. A21. NGC 4473: diE. Positions in ′′. N at 12 o’clock.B −R isochromes: 1.575− 1.675 (0.025). B isophotes: 16.75−18.75 (0.5). The flattening of the isochromes, and their shiftalong the minor axis, suggest dust concentration in the disk

Fig. A22. NGC 4478: boE. Positions in ′′. N at 12 o’clock.B −R isochromes: 1.575− 1.675 (0.025). B isophotes: 17.25−18.25 (0.5). This boE, a member of the family of rotationallysupported minor objects, has no significant colour pattern, be-sides a rather sharp central red peak

Fig. A23. NGC 4494: unE. Positions in ′′. N at 12 o’clock.B−R isochromes: 1.45−1.60 (0.025). B isophotes: 17.0−19.0(0.5). This slightly disky object shows no obvious colour pat-tern, but for a sharp nearly cemtral red peak. HST V-I imagesdisclose a small dust ring (see text)

Fig. A24. NGC 4564: diE. Positions in ′′. N near 2 o’clock B−Risochromes: 1.60− 1.70 (0.025). B isophotes: 17.2− 19.2 (0.5).Poor S/N . Possible faint, near central, colour pattern. Thedisk is clearly seen farther than 10′′ along the major axis, withpossible dust concentration (see Paper II)


Fig. A25. NGC 4621: diE. Positions in ′′. N near 8 o’clock.B −R isochromes: 1.595− 1.695 (0.025). B isophotes: 16.75−18.25 (0.5). Poor S/N . The flattening of the inner isochromessuggests some dust concentration in the disk

Fig. A26. NGC 4649: boE. Positions in ′′. N near 1 o’clock.B−R isochromes: 1.655−1.665 (0.01). B isophotes: 17.0−19.0(0.5). Poor S/N . The colour distribution of this flat core galaxyis also remarkably flat. See Fig. 9

Fig. A27. NGC 4660: diE. Positions in ′′. N near 1 o’clock.B−R isochromes: 1.55−1.65 (0.025). B isophotes: 17.0−19.0(0.5). The flattening of the inner isochromes suggesgts somedust concentration in the disk, while their tilt might indicatea faint dust lane crossing the core

Fig. A28. NGC 5322: boE. Positions in ′′. N near 3 o’clock.B−R isochromes: 1.45−1.55 (0.025). B isophotes: 17.0−19.0(0.5). The flattening of the inner isochromes suggests some dustconcentration in the inner “decoupled” disk. See also the HSTframe in Lauer et al. (1995)


Fig. A29. NGC 5576: boEp. Positions in ′′. N near 10 o’clock.B−R isochromes: 1.425−1.55 (0.025). B isophotes: 17.0−19.0(0.5). Poor S/N . No colour pattern is seen, but for the rathersharp central peak

Fig. A30. NGC 5813: unE. Positions in ′′. N near 2 o’clock.B−R isochromes: 1.60−1.70 (0.025). B isophotes: 18.0−20.0(0.5). Poor S/N . An asymmetric core dust pattern is present

Fig. A31. NGC 5831: diE. Positions in ′′. N near 3 o’clock.B−R isochromes: 1.61−1.71 (0.025). B isophotes: 17.4−19.4(0.5). Poor S/N . An asymmetric core dust pattern is present,again with some indication of flattening of the isochromes

Fig. A32. NGC 5845: diE. Positions in ′′. N near 3 o’clock.B−R isochromes: 1.60−1.70 (0.025). B isophotes: 17.0−19.0(0.5). Poor S/N . The flattening of the inner isochromes sug-gesgts some dust concentration in the inner “decoupled” disk.See also the HST frame in Lauer et al. (1995)

Documents

Colour distributions in E-S0 galaxies · 246 R. Richard: Colour distributions in E-S0 galaxies. IV. deemed useful to analyse the best B and R frames in Nieto’s CFHT collection,